NC Bench
Overview
About
Tests
Models
Model
Gemini 2.5 Flash Lite
Test
Dialogue tags
Scenario
Write 200 words with 90% dialogue
Run ID
2025-11-18-01KACP4SPHKMPAYNNC87N8E054
Model input
User/Human
You are a author of comedic heist books. Please write a 200 word scene that has 90% words of dialogue between two thieves named Molly and Bill who are planning to break into a children's party to steal all the presents. Only include the prose of the scene without any markdown or commentary.
Model response
Messages
Meta
AI
"Right, the cake is the decoy," Molly whispered, flicking a miniature plastic dinosaur under a table. "Decoy? For what?" Bill hissed, adjusting his oversized clown nose. "To distract the tiny humans. While they're busy shoving frosting in their face, we grab the loot." "And the loot is…?" "Every single gift. Think of it, Bill! Imagine a pile of unopened joy, ripe for… redistributing." "Sounds like a lot of tiny hands reaching." "That's where you come in. You're the happy distraction. Juggle. Tell bad jokes. Make balloon animals that look vaguely threatening." "My balloon dachshunds are legendary!" "Exactly. Meanwhile, I'll be the silent shadow. The phantom of present peril. I'm thinking a black leotard and maybe some strategically placed glitter." "Glitter? Molly, it's a kiddie party, not a rave." "Details, Bill, details. The point is, we get in, we grab, we get out. Before the parental units realize their precious progeny's joy has been… rehomed." "What if they catch us?" "Then we blame it on the sugar rush and a rogue unicorn." "A rogue unicorn? You're a genius, Molly. A slightly unhinged genius with excellent taste in leotards."
created_at
"2025-11-18T23:52:36.926783+00:00"
Tuesday, November 18th, 2025 at 11:52:36 PM GMT+00:00
model
"google/gemini-2.5-flash-lite"
app_id
182717
external_user
(null)
streamed
true
cancelled
false
latency
287
moderation_latency
(null)
generation_time
1922
tokens_prompt
68
tokens_completion
275
native_tokens_prompt
67
native_tokens_completion
294
native_tokens_completion_images
0
native_tokens_reasoning
0
native_tokens_cached
0
num_media_prompt
(null)
num_input_audio_prompt
(null)
num_media_completion
0
num_search_results
(null)
origin
"https://ncbench.com/"
is_byok
false
finish_reason
"stop"
native_finish_reason
"STOP"
usage
0.0001243
api_type
"completions"
id
"gen-1763509954-K4Rym33UsEMF7Dm6YGVo"
upstream_id
(null)
total_cost
0.0001243
cache_discount
(null)
upstream_inference_cost
0
provider_name
"Google AI Studio"
Evaluation details
Result
Evaluator
Details
Meta Data
75.1556%
Matches word count
n/a
neededClean
false
words
187
99.9932%
Dialogue to Total Word Ratio
Ratio: 90.91%, Deviation: 0.91%
neededClean
false
wordsTotal
187
wordsDialogue
170
87.5744%